Github Workflow

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed GitHub integration, but it can make changes through a linked GitHub account, so users should install it with care.

Before installing, check that you trust MorphixAI and the linked GitHub account. Use least-privilege repository access where possible, verify the active account before acting, and require explicit approval before creating issues or pull requests, closing issues, or triggering workflows that may run CI/CD or deployments.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill documents state-changing GitHub operations such as create_issue, update_issue, create_pull, and especially trigger_workflow without clear warnings, confirmation guidance, or discussion of repository/CI side effects. In an agent setting, this increases the risk of unintended writes, workflow execution, deployment initiation, or consumption of CI resources when a user asks for read-like GitHub help but the agent chooses a dangerous action.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal